Should the Bar “Bar” AI? Use in Legal Profession Has Risks and Rewards

July 13, 2023

| Share

Do you want DE Insights Delivered to Your Inbox? Sign up Today!

ChatGPT, a platform from OpenAI, has been leading the charge for general use Artificial Intelligence (“AI”), with serious implications for businesses. AI is, at its core, “… the overarching description for technologies that use computers and software to create intelligent, humanlike behavior.”[1] While the legal profession can be slow to adopt new technologies, the industry has seen an impressive, if, albeit dramatic, push to adopt AI technology — and, legal clients may be expecting their counsel to do the same. According to a Thompson Reuters Institute report, more than 80% of law firm leaders believe AI can be applied to legal work now and more than 50% believe it should be applied to legal work.[2] AI is currently being used in reviewing contracts, the discovery process, legal research, and even litigation strategy.[3] However, before legal professionals can truly evolve with AI, practitioners (and their clients) need to understand the risks and rewards that come with adopting these new technologies.

Risks of AI Use

In February 2023, a Defendant was planning on using AI to try to battle a traffic ticket issued in California.[4] DoNotPay, a New York-based company, had been using ChatGPT and other AI sources to provide written defenses to traffic tickets on behalf of clients. The Defendant intended to use smart glasses in court and then respond to any questions with AI generated arguments through an earpiece. Once word got out of the plan, state bars sent DoNotPay threats for unauthorized practice of law without a license, a misdemeanor in several states. Id. Though DoNotPay ceased its efforts, this did not end the incursion of use of AI in the legal field.

In May 2023, plaintiff’s counsel in Mata v. Avianca Inc., Case No. 22-CV-1461 (S.D.N.Y., filed Feb. 22, 2022), were called out by opposing counsel and Judge Kevin Castel that the cases cited by plaintiff’s counsel appeared to be “bogus.”[5] Counsel had used ChatGPT to research cases, including case citations and quotes, all of which were fake and fabricated by AI. After a hearing on the manner in how such cases made their way into a Federal Court filing, the Court issued sanctions against the counsel, including fines of $5,000 each.[6] This should serve as warning against unchecked use of AI in court.

In Federal courts, counsel are required by Federal Rule of Civil Procedure 11 (popularly known as “Rule 11”) to sign off on all documents filed with the Court. An attorney’s signature under Rule 11 acts as certification that everything in the filing is true and correct within the best of that attorney’s knowledge and belief.[7] This includes anything put in front of the Court, not just evidence but also the legal authorities and arguments put forth. In addition to Rule 11, individual judges have also started issuing orders requiring all attorneys appearing before them to disclose any use of AI and to certify that human review was conducted on any AI generated content.[8]

While AI can be useful in drafting certain documents, caution is advised regarding the reliability of AI generated content. OpenAI has the following in its ChatGPT Terms of Use:

Accuracy. Artificial intelligence and machine learning are rapidly evolving fields of study. We are constantly working to improve our Services to make them more accurate, reliable, safe and beneficial. Given the probabilistic nature of machine learning, use of our Services may in some situations result in incorrect Output that does not accurately reflect real people, places, or facts. You should evaluate the accuracy of any Output as appropriate for your use case, including by using human review of the Output.[9]

Accordingly, OpenAI disclaims the accuracy of the information it provides. While most companies have a blanket indemnification in their terms of use, the specificity of OpenAI’s “Accuracy” section should give pause to any attorney seeking to use the platform as a resource.

Attorneys must also consider client-confidentiality when using AI. Inputting client-specific facts into an AI tool could cause data mishandling and a possible breach of client-confidentiality. Each AI tool uses its own servers, outside a firm’s infrastructure, to provide the digital analysis requested. The security of these servers is unknown. Relinquishing control over client data must be a factor in considering the use of AI in a legal practice.

Rewards of AI Use

Ultimately, the greatest benefit that AI provides attorneys is time savings. However, AI is not limited to quickly creating output for an attorney. There are a plethora of AI tools that can assist in the day to day administrative tasks that often burden attorneys and the judicial system. For example, Gideon is an AI chatbot that can help streamline client-intake questions.[10] Another AI tool, Diligen, assists attorneys by reviewing contracts for specific clauses or changes in a document and can provide a summary of lengthy documents.[11] Courts in New York and California have used the Correctional Offender Management Profiling for Alternative Sanctions (“COMPAS”) system, an AI tool that assesses the likelihood a defendant is to reoffend, in bail determinations.[12]

AI can also help in several other areas of daily legal work such as analysis of discovery documents for relevant or confidential information, or proofreading briefs.[13] What could be hours of legal billing to find the proverbial needle in the haystack can be reduced with a few clicks of the mouse, saving the client legal fees and providing the attorney additional hours of capacity for other tasks. AI also provides attorneys with the ability to assess voluminous documents at the onset of a matter, reducing the amount of time needed to accurately understand the full breadth of the matter from the beginning of an engagement and analyze potential risks and outcomes, rather than sift through pages to establish timelines or sequences of events.

Conclusion

Ultimately, AI further allows attorneys to pass on savings to their clients and improve their client relationships — first, by reducing the number of hours necessary to complete time-consuming tasks, and second, by allowing attorneys to time to tackle the matters and/or issues that cannot be handled by AI, such as signing under penalty of perjury. However, both attorneys and their clients should be wary of over-relying on AI. For now, AI can’t guarantee the accuracy of its byproduct (or interface with clients). But these technologies continue to evolve, and we may need to reconsider our position, and reliance, on AI, in the near future.

——————————————————————–

This DarrowEverett Insight should not be construed as legal advice or a legal opinion on any specific facts or circumstances. This Insight is not intended to create, and receipt of it does not constitute, a lawyer-client relationship. The contents are intended for general informational purposes only, and you are urged to consult your attorney concerning any particular situation and any specific legal question you may have. We are working diligently to remain well informed and up to date on information and advisements as they become available. As such, please reach out to us if you need help addressing any of the issues discussed in this Insight, or any other issues or concerns you may have relating to your business. We are ready to help guide you through these challenging times.

Unless expressly provided, this Insight does not constitute written tax advice as described in 31 C.F.R. §10, et seq. and is not intended or written by us to be used and/or relied on as written tax advice for any purpose including, without limitation, the marketing of any transaction addressed herein. Any U.S. federal tax advice rendered by DarrowEverett LLP shall be conspicuously labeled as such, shall include a discussion of all relevant facts and circumstances, as well as of any representations, statements, findings, or agreements (including projections, financial forecasts, or appraisals) upon which we rely, applicable to transactions discussed therein in compliance with 31 C.F.R. §10.37, shall relate the applicable law and authorities to the facts, and shall set forth any applicable limits on the use of such advice.

[1] How Artificial Intelligence is Used in Legal Practice | Bloomberg Law

[2] Generative AI in law firms: For many, such technologies are still a great unknown | Reuters

[3] AI in Law: The Transformation of Legal | Clio

[4] An AI robot lawyer was set to argue in court. Real lawyers shut it down. : NPR

[5] Lawyer apologizes for fake court citations from ChatGPT | CNN Business

[6] cnbc.com/2023/06/22/judge-sanctions-lawyers-whose-ai-written-filing-contained-fake-citations.html

[7] Federal Rules of Civil Procedure Rule 11. Note that most state courts have an identical requirement.

[8] https://www.reuters.com/legal/transactional/another-us-judge-says-lawyers-must-disclose-ai-use-2023-06-08/#:~:text=(Reuters)%20%2D%20A%20judge%20on,concerns%20related%20to%20confidential%20information.

[9] Terms of use (openai.com)

[10] https://www.gideonlegal.com/

[11] https://www.diligen.com/

[12] https://mashable.com/article/donotpay-artificial-intelligence-lawyer-experiment